A guide to behavioral analytics
What is behavioral analytics?
Behavioral analytics is a tool that reveals the actions users take within a digital product. It captures and organizes raw event data into a timeline for each user’s behavior, also known as a user journey. Teams use behavior analytics to determine what users like and don’t like and, by inference, what adjustments can make the product more valuable.
Why should companies use behavioral analytics?
Most product, marketing, and analytics teams live in constant pursuit of the question, “How are customers using the product, if at all?” Behavioral analytics software provides concrete answers with a visual interface where teams can segment users, run reports, and deduce customers’ needs and interests.
Without behavioral analytics, teams are stuck using insufficiently detailed demographic data and so-called vanity metrics. As Streaming, Sharing, Stealing co-author Michael D. Smith explained to The Signal, if a company wants to personalize its service to users, it needs their behavior data. A streaming movie platform can’t know that a user loves horror films, for instance, simply based on their age, gender, or nationality.
Behavioral analytics can provide user-level data so teams can answer questions like:
- What are users doing within the product?
- Where do users get stuck?
- How do users react to feature changes?
- Is the product as simple and intuitive as it could be?
- How often do users reach profitable outcomes?
- How can the team nudge users to be more successful?
Conducting behavioral analysis is more complicated than simply running reports in the analytics tool. “Analyzing generic data doesn’t magically produce answers to unidentified problems,” wrote Drew Hendricks, a technology writer for Inc. Teams must first identify what they want to achieve and write down the paths they expect users to take. Only with preset expectations can teams identify whether users are deviating from the ideal path and redirect them.
The tech-driven insurance provider Lemonade, for instance, adjusted its user paths to increase revenue. The team knew their goal was to convert more website and app visitors into paying customers and with Mixpanel analytics, they noticed a “staggering drop-off” in user flow right before the point of purchase. By analyzing the page where the drop-off occurred, the team realized it was due to a technical error and a weak call-to-action (CTA). They fixed the bug and reworded the CTA, which led to a 50 percent increase in the number of users who purchased additional coverage.
Why behavioral analytics is different
What sets behavioral analytics apart from other types of analytics is that it combines two technologies: user segmentation and event tracking. While some analytics vendors only offer one or the other—user data or event data—behavioral analytics unites the two for a complete customer view. It ties users to the events they trigger to produce a map of their actions, also known as a user flow, or customer journey.
Viewed either alone or in aggregate, user journeys tell stories that teams can use to tweak and improve their product development, marketing, and launch strategies.
Steps to successful behavioral analysis
Customer behavioral analysis requires careful planning and each team’s success is a function of how carefully they implement the analytics tool and how seriously they take their tracking plan.
Behavioral analysis is not a race. The first half of the implementation process should be spent planning and all teams who will eventually benefit from user intelligence need to have a hand in selecting and deploying the tool.
Teams can prepare themselves to conduct behavioral analysis in five steps:
1. Select goals, KPIs, and metrics
To determine whether users are reaching the right goals, such as driving revenue for the business, teams must select the KPIs and metrics that indicate progress toward those goals.
A fitness app, for instance, could track paid subscribers and subscriber growth because it makes money through monthly subscriptions. An enterprise resource planning (ERP) software, on the other hand, could track users that complete their onboarding sequence because that factor is deterministic of a second-year renewal. Teams should record their metrics.
2. Define the most desirable user journeys
Based on the service or app’s design, what are the most common paths for users to reach their goals? If the product has already been launched, teams can use actual user data to answer this question. If the product is pre-launch, the team can use the design team’s wireframes of the intended flow.
All user journeys, or paths through the product that the team tracks, should end in some type of a desirable outcome for both the customer and the business. An e-commerce website, for instance, could track a user from a referred source to a page visit, adding an item to their shopping cart, and then checkout, because that flow leads to purchases. Alternatively, a streaming music app could track users as they move from its homepage to playing a song and, hopefully, purchasing that song.
3. Create a tracking plan
Based on the presumed user flow, teams can decide which events they’ll need to track within the product. It can seem appealing to track everything but this is a mistake—too much data can clutter the analytics and make information more difficult to find. Track events and users based on whether the data will be useful.
Some events have, within them, multiple properties. The team for a music app, for instance, could decide that the event for playing a song should have nested properties such as the song title, genre, and artist. To keep events and properties organized, companies typically create a tracking plan in a spreadsheet. This acts as a directory of all events and serves as a map for implementing the analytics tool.
A tracking plan is a mutable document that will be revised and updated as the product, team, and goals change. To reduce the burden of trying to share and control access to the spreadsheet, Mixpanel offers a feature called Lexicon which stores the event name taxonomy within the analytics for all to see.
Involve all teams—analytics, product, marketing, and engineering—in drafting the tracking plan. Members of each will need to understand how the users and events are named and organized if they’re going to run reports and understand the results.
4. Set a unique identifier
Most digital products today exist across multiple platforms and this makes it difficult to track unique users. One user can appear to be multiple people unless assigned a unique identifier—either an email or string of characters—that persists across platforms and devices and connects the customer journey. Teams should ensure their behavioral analytics platform vendor provides a unique identifier and that it won’t change over time.
5. Implement analytics and begin event tracking
Once the tracking plan is complete, companies can deploy behavioral data analytics software and use its SDK or API to integrate it with their products. That’s when they assign user IDs and set up user and event properties as outlined in the tracking plan.
It’s not uncommon for teams to discover additional events they want to track during implementation. This isn’t an issue as long as they update both the tracking plan and the analytics service.
Before the production level event tracking goes live, teams should use test devices to verify the event and user tracking is firing properly. Once working, teams are ready to begin analyzing their users’ behaviors.
How to apply the results of behavioral analytics?
Most teams first try to understand who their users are with segmentation, which allows them to separate users based on characteristics and behaviors. An e-commerce app, for example, could segment for recent users who added items to a shopping cart but then abandoned. Or, they could filter for power shoppers who access an app multiple times a day.
Segmentation allows teams to learn about their users to build more complete customer profiles. They can save user segmentations, known as cohorts, and make adjustments to their product and marketing to make it more profitable with each segment.
Media and entertainment company STARZ PLAY, for instance, segmented users that signed up through its free trial offer and found that some users were gaming the system for multiple trials. By creating alerts for the negative behavior, the product team closed the loophole and saved 8x on its marketing spend.
Teams can track users’ progress toward outcomes such as purchases or signups with funnel reports. Funnels display a series of stages in a user journey, as well as how many users are progressing from one stage to the next. A fitness app could use funnels to see how many users progress from download to signup and purchase. If one stage has a low conversion rate, it’s a signal that that stage needs attention.
Teams can also deduce which behaviors are correlated with high retention. A media site, for instance, could look at the cohort of users that continue to return to the site eight weeks after signup to see if they share certain behaviors, such as a propensity to leave comments on articles.
With a view of what’s happening within the product, teams can run experiments and make alterations to improve the value users get from the product.
How to choose the right behavioral analytics provider
The best behavioral data analytics solutions are widely compatible, offer quick results, and are versatile. They allow teams to integrate their full range of digital products, from mobile and desktop apps to internal services like the CRM and customer support system. They have friendly, intuitive interfaces that allow teams to quickly find quick answers, and they provide a breadth of functionality, including some that teams don’t yet need, but may in the future.
Swapping out a behavioral analysis platform can be a costly and time-consuming endeavor. Teams can save themselves time and money by evaluating vendors with great caution and investing in a platform that offers them room to grow into.
Look for a behavior analytics solution with:
- The ability to automatically capture user and event data
- The ability to access data quickly and query it in a variety of ways
- Pre-built reports such as funnels, cohorts, and retention
- Guide rails, automated notifications, and recommendations
- A visualization component